note: the left scale is a log scale
----
i've made several changes and added a second chart whose position (a bit up
and to the right of the original chart) is aligned so that its Y axis % scale aligns
with the original Y axis USA population scale.
(note: i don't know how to make a chart with two Y axis' scales in mac's numbers
- this is my way of faking it)

the new chart also has two CFR (Case Fatality Rate) plots. the red one is the
deaths divided by the cases on that date & the other [CFR9] is deaths divided
by the cases on the date 9 days earlier. the 9 days was gotten from the time
between the two green vertical lines described below ----

----

note: as i change the plot the colors change so the below's colors might not
be correct but i hope you get the idea.

i have added past today's date a linear projection of the deltas in bold lines. this
might prove useful since the exponential growth period seems to be over and the
day to day growth seems to be leveling off.

=

i've also added two green vertical lines at 3/20 and 3/29. these dates were choosen
because they represent the first significant peak in the deltas of cases and deaths.
the implication to me is that this might be close to the average time delay between
showing up on the cases chart and the resulting change in the death chart sometime
later. this ia about 9 days. in other words their might be, on average, 9 days between
whatever it takes to show up on the cases chart and dying.

an implication of this might be the approximate death rate is the number of deaths
divided by the number of cases 9 days earlier. for example at the first vertical green
line there are about 20,000 case and at the second vertical green line there are about
2,000 deaths. using this method, the death rate is 10% (2,000/20,000).

== the below may be out of date ==

the above USA covid19 chart explained top to bottom:

the heavy red part from 3/1 to 3/18 is the data is used to extrapolate
the exponential growth of USA cases. this is plotted as the orange straight line
continuing on from the heavy red line.
the wavy blue line is the actual number of covid19 USA cases.

the next line down is the delta, ie, the difference of that day from the day
before. the straight orange line is the projection delta and the wavy green
line is the actual delta. as you can see the exp growth has slowed from 3/26
which is good news.

the third line down is the USA covid19 deaths. the heavy section is what i used
to extrapolate the the deaths assuming exponential growth.
the wavy red line is USA death data and the blue straight line is the extrapolated exp
deaths.

the fourth line is the delta of change of today's data from yesterday'a data with
the straight line being the extrapolated delta of deaths and the wavy line the actual delta
of deaths.

==========

while the exp growth in cases tapered off starting 3/26 it looks the the exp growth in deaths
might have started to to taper off on 4/2 - i need more data to verify this is real.

----

btw, the extrapolated doubling time (2X) for cases is about 2.5 days and the 10X time